Sr. Principal High-Performance Computing Architect

Company: Tenstorrent Inc
Location: Santa Clara
Posted on: November 1, 2024

Job Description:

We are seeking a skilled and experienced High-Performance Computing (HPC) Architect to join our team. As an HPC Architect, you will play a crucial role in designing and implementing cutting-edge architectures for high-performance computing systems, enabling efficient and scalable computation for scientific, research, and data-intensive applications. You will collaborate closely with cross-functional teams, including hardware engineers, software developers, and domain experts, to deliver optimized solutions that meet the demanding requirements of HPC workloads.
Responsibilities

• 
  
• Design and develop high-performance computing architectures that deliver exceptional computational performance, scalability, and energy efficiency.
  
• Collaborate with hardware engineers to design and optimize the system's computational components, including processors, accelerators, interconnects, and memory subsystems.
  
• Work closely with software developers to define and implement software frameworks, libraries, and tools that maximize performance and productivity on the target HPC architecture.
  
• Conduct performance analysis, benchmarking, and modeling to identify performance bottlenecks, optimize system parameters, and guide architectural enhancements.
  
• Define system-level requirements, including processing power, memory capacity, I/O bandwidth, and storage capabilities, and ensure compliance with industry standards and customer expectations.
  
• Evaluate and select appropriate technologies, including processors, accelerators, and network fabrics, based on application requirements, performance characteristics, and cost considerations.
  
• Collaborate with domain experts and application developers to understand the unique requirements of specific scientific, research, or data-intensive workloads and propose tailored architectural solutions.
  
• Stay up-to-date with the latest advancements in HPC hardware, software, and industry trends to guide future system designs and technology adoption.
  
• Work closely with vendors and suppliers to evaluate emerging HPC technologies, assess their suitability for specific applications, and negotiate supply agreements.
  
• Provide technical guidance and mentorship to junior team members, fostering knowledge sharing and best practices within the HPC architecture domain.
  
  Experience & Qualifications
  
  • 
    
  • Bachelor's or Master's degree in Computer Engineering, Electrical Engineering, or a related field. A Ph.D. is a plus.
    
  • Strong expertise in high-performance computing architecture design, including processors, accelerators, interconnects, and memory subsystems.
    
  • Proven experience in designing and optimizing HPC architectures for scientific, research, or data-intensive applications.
    
  • Proficiency in parallel programming models and frameworks, such as OpenMP, MPI, CUDA, or OpenCL, and their application to HPC workloads.
    
  • Solid understanding of performance analysis and optimization techniques for parallel computing, including profiling, tracing, and performance counters.
    
  • Familiarity with industry-standard interconnects and network fabrics, such as InfiniBand, Ethernet, or Omni-Path, and their impact on HPC system performance.
    
  • Knowledge of memory subsystems and memory hierarchy designs, including cache coherence protocols, memory models, and NUMA architectures.
    
  • Experience with HPC software stack components, such as compilers, runtime systems, job schedulers, and scientific libraries.
    
  • Strong programming skills in languages commonly used in HPC, such as C, C++, Fortran, or Python.
    
  • Excellent problem-solving abilities and the ability to analyze and address complex performance and scalability challenges.
    
  • Strong communication and collaboration skills to work effectively with cross-functional teams and domain experts.
    
    Job Type:
    This role is hybrid, based out of Santa Clara, CA.
    #J-18808-Ljbffr

Keywords: Tenstorrent Inc, Rancho Cordova , Sr. Principal High-Performance Computing Architect, Professions , Santa Clara, California